From the time you begin prototyping, to your finished product, most of us will have to rewrite either the entire game or significant parts of it. Especially if you're diving into some new technology, the techniques you learn along the way can make your early work pretty miserable, and, worst case scenario, cause some problems with the game itself.
There's probably a few developers whom have been at this for a while and are just honing their craft. This post isn't for them. Getting into game development comes with many challenges, and unspoken phases. One of them: The Rewrite.
There's a fairly large cost to benefit analysis I automatically jumped to when considering where in my road map I was at first. Do it too early, and I'll continue learning things and end up with artifacts that I'll have to redo anyway. Do it too late and I'll have wasted so much time working around tech debt.
This is a BIG thing in other parts of human civilization, namely the Wait Calculation for interstellar travel.
...This is because a slow spacecraft would probably be passed by another mission sent later with more advanced propulsion.
Basically, how long are we willing to wait before we're reasonably certain that our personal progress in the future won't hinder our efforts today? I spent a lot of time looking at the road map, making adjustments, moving things around... half I justify for the sake of the community. Half I justify for my own sanity. But a full rewrite? Months of no progress, and nothing to show.
What I did
Arms of War doesn't have a huge player base right now, so that now isn't a bad time to detour from releasing features or improvements. Although I fought to squeeze in one feature that was in the Beta stage of AoW (currently in Alpha): Skirmishes.
Skirmishes is a second game mode, meant to be very fast-paced and effect Battles in a fun way. Had I dropped out and started the rewrite prior to it's release, the few die-hard members of my community would have left because there's simply nothing to do other than the same limited game over and over. With a second game mode, atleast they'd be satiated for the duration of the rewrite.
I chose to invest time into a feature that wasn't planned until the middle of the next version, and even then after a host of others. It was important to give something substantial to the small community before undertaking a lengthy journey as a reason to stick around and see what comes out the other side of the rewrite.
There was a second reason to jump-start the rewrite so early as well: I had brought on another very talented developer and someone to help with the administrative side of the house. With the rush of extra resources, the time taken for the rewrite is lessened. Bonus! And, what better way to learn how a program works than to remake it in a better image? This will help with the spin-up time for the new development crew. I placed a lot of leverage in this category, but even with all that, I pushed Skirmishes anyway.
For every bit you ask of your community, you've got to give them a reason for it.