"Perfect is the Enemy of Good" - Voltaire

One of the things that's fascinating about design is that it's a series of balancing acts. You absolutely can spend until the heat death of the universe creating the most streamlined, perfect, extensible and human readable piece of code. Definitionally, you will never create a product. On the other hand, not planning at all will yield an incomprehensible mess that leads to lack of optimization, confusing decisions, technical debt and more. This path is only marginally more acceptable because it can still lead to a product. The companies that do publish software in this domain are what are referred to often as Shovel-ware or Abandon-Ware.

So one of the ongoing philosophical discussions I have with myself while I slave away in the Code Mines is exactly where to strike that balance. In a lot of ways, putting yourself into a design document can be a convenient excuse to not Do The Hard Thing.

One of the lessons learned from Mark Brown (Game Maker's Toolkit) is the principle of prototyping and focusing on making the gameplay fun before getting too far down the road of complex story beats or immaculate pixel art. 

One of this week's projects was overhauling the dialogue and social system and building the seed's for it's further extensibility. The other was building what I've phrased as the MkII assets. In building tilemaps and ruletiles I am building the inevitable workflow and once the system is in place I can create a middle term solution to add slightly more polish when I start handing out my vertical slice so it looks like a game (even if it is a crappy one) but also forms the bedrocks for tools that should only require a simple switch out

Cutting and Running

"Americans spent millions of dollars making a pen that would work in space. The Soviets used a pencil"

There's also something to be said of when it is time to abandon a technical solution which *could* be a time saving measure. I wanted a "death animation" similar to how final fantasy monsters fade out and disappear. I had a "brilliant" idea that would overlay a filter that was essnetially static and then increasingly zero out the alpha of the pixels of the sprite based on the pattern set by the static overlay. I couldn't get this to work and I realized that both the perspectives of human and computing resources, I could probably come up with a better solution to that, which was creating a file in GIMP that creates five increasingly transparent, decreasingly saturated sprites with pieces missing. They just end up on the sprite sheet. A relatively light piece of work (especially now that the workflow is determined), but actually increases the readibility and reduces the overhead. 

Now, the canard about the Space Pen overlooks the fact that there are some pretty good reasons not to use something that sheds highly conductive material in a zero G environment, but the feeling behind the (sort of erroneous) story is as follows:

"The Simple Answer Can Be Better, Actually"