I donate a lot of my free time to MIT licensed open source projects. Primarily MeerK40t, pyembroidery, and svgelements.

I focus on reducing the amount of money people spend on software and hardware. I often work on projects where I feel costs outweigh the efforts made. This often leads me towards projects and in directions of lesser served niche projects. So, I don't think raster preprocessing scripts, expensive replacement laser-controllers, or embroidery software are worth as much as people pay.

I've incurred a few costs myself usually in the low tens of dollars in these projects such as buying hardware, some accounts for signing/publishing software, and there's also the thousands of hours of skilled labor. Some testing and advancements are not possible without physical hardware to test it on, and usually that costs money.