sbonner wrote:So, the question is, how can I make it economically justifiable?
It's an excellent question, (and the distinction between repair / upgrade vs. make is absolutely valid) but I'm afraid I don't have a particularly satisfying answer overall. I could try to argue that it's an investment as you go and that over time, the accumulated knowledge and catalogue of bits and bobs could end up cost-effective, but I really do think that's a fairly weak argument in the short view. You wind up with hand-made (and frequently shoddy... and sometimes failed) items that you could probably get cheaper and faster from a cart in the mall.
Simply put, I'm at a point in my life where I'm comfortable looking at this as an entertainment expense. I'm just making toys, not because it's in any way cheaper or more effective than buying them, but because I enjoy the process of figuring out how to make them. I realize that not everyone is so fortunate (and we've all been in that position, too). The MAME box, for example, is made up of a $40 computer, a $15 joystick and a couple of $2.50 buttons... and you could probably find an unlicensed joystick that plugs into your TV and does the same thing for $20 or so.
There are some compelling counterpoints, at least. I can take that arcade machine apart and reuse every single piece of it - down to the wires themselves - elsewhere. I could hook it up to an ethernet cable and cell phone charger and repurpose it as a generic desktop or home server. $65 for an arcade machine + $0 for a server is arguably approaching reasonable. On a personal level, I also feel slightly better about doing so than buying intentionally disposable goods but the experience gained in the process is the real reason.
All I've got are a few things to try to mitigate the issue, if not really resolve it.
1. Buy the generic components in bulk. Different color buttons are different SKUs (and therefore don't get you the bulk rate!)
2. Find a local source. Save on shipping. Ask staff questions so you buy the right stuff once.
3. Build everything to take it apart. Everything is a prototype and as such can be disassembled and reused.
4. Model in your head first. Right now, as I type this, I'm going through how to use the fewest pre-crimped wires for the 8 buttons on the next version... and the parts haven't even shown up yet!
5. Consider some purely digital projects. Interested in learning a programming language, trying out some web development, or finding your way around a linux box? A $30 Raspberry Pi A could be used as a development machine that is pretty much bulletproof. If you screw it up, all it takes is time to wipe the card and reinstall. Even cheaper, set up a VM in Virtualbox and slap a copy of Mint or Ubuntu on it.
6. It'll get better over time! Not only will your stock of leftover / reusable bits build up, but you'll have a better idea of what you'll really need and avoid needless expense. Not helpful in the short term, but it's a valid long-term goal, I think.
As a side note, I can't stop raving about the Pi, though. $30 or 40 and you have a computer that's made to have random wires hooked up to it, which is brilliant. Imagine running bare wires from a LED onto some random pin on your motherboard and then trying to figure out how to make your computer tell it to light up. You're supposed to do that with the Pi, which is either insane or the only sane thing in an insane industry - I'm not sure which.
I'd consider that a rare case of a component that's worth as much or more than the result for which it's used. Arduinos are slick, but absolutely for plugging wires into it... and not much use without the additional expense of things on the end of those wires.
Free OS in a free VM lets you make stuff that's purely digital, for free! And as you get ideas to start integrating real-world stuff, then you start looking at hardware with that experience behind you and you'll have a solid foundation on the software end and can focus on the physical components.
I have no idea if that's helpful or not. Fingers crossed.