I ran across an old email exchange with Curt Beckmann in which he said something that made me stop and think: "It's not 100% clear to me that software developers are appropriate technologists, or vice versa."
I think this is a key point in understanding why the open source appropriate technology - or open source engineering, for that matter - hasn't taken off yet. One huge reason the open source software movement was able to bootstrap so rapidly was that it was inherently about people eating their own dogfood and solving their own problems; in other words, coders were writing code for themselves, code to help them code better.
Similarly, I've seen my fellow engineers whip out custom grips, tools, clamps, molds, bits of debugging firmware, hacked-together circuits, etc. to make their design and prototyping work go more smoothly. You use the tools of your trade to bootstrap your work in your trade. That's only natural, and if those hacks are easily shared, they compound exponentially.
But for an appropriate technologist - let's say a sanitation engineer - to work with code, that's a different medium entirely; writing code won't typically directly aid you in building a composting toilet or whatever. It's one step more removed, and that barrier is a huge gulf. (For the same reason, I'll hazard that this is why most coders don't like writing documentation. It usually doesn't get the code to be any more functional.)
Might this also be why there's so little good software for electrical engineers out there? And why nearly all the decent software for mechanical, biological, materials, etc. engineers is made either by big corporate entities (with widespread usage) or grad students (as individual script hacks that tend to languish in obscurity)?