On Friday I reinstalled Red Hat 9 on my computer. On Saturday I figured out why I couldn’t build Dillo on the virtual Conectiva system (the only reason I tried to install the real thing). On Monday I made a remark on the Dillo mailing list that, after trashing the system trying to install Conectiva, “unless/until I can set up a spare system solely for trying things out, anything else… will run under User-Mode Linux.”

And that got me thinking.

A spare system wouldn’t need to be elaborate. I wouldn’t be playing games on it. I wouldn’t be doing graphics work on it. I wouldn’t even be doing web development, word processing, or checking my email. Most of the time it wouldn’t even be running – just when I wanted to try something new, or when a new release of Dillo came out and needed RPMs. And since I have a spare KVM switch, I’d only need to find space for the case, and wouldn’t need to worry about a monitor.

I’ve mentioned before that I’ve upgraded my computer piece-by-piece since 1994. Well, when you do that, you end up with a lot of spare parts left over. Sure, they’re older, slower, smaller, etc. than what you’ve got now, but if they worked when you took them out, they probably still work now. I’ve been meaning to go through all the boxes, cards, drives etc. and get rid of things I don’t need anymore, but I’d never gotten around to it. Well, on Monday I finally had motivation.

I went through looking for parts I could use to put together a spare, expendable system – one where it wouldn’t matter if the entire hard disk got wiped. I found three hard drives (two of them too small to be useful), several sticks of RAM, network and video cards, and a motherboard and a CPU that wouldn’t fit together. That left: a case, a CD-ROM, and either a CPU to go with the motherboard, or a motherboard to go with the CPU.

It turned out my boss was getting ready to throw out some old equipment, including a huge mega-tower with a 450 MHz K6-2. The motherboard I have used to hold a K6-2. (Where it is now, I have no idea – I don’t think it’s the one I fried, especially since the motherboard seems to work.) So now I had a processor. My parents had recently replaced an extremely flaky computer, so I got a CD-ROM from that. Then I went to Fry’s and picked up a $30 case and $13 floppy drive.

That’s right: I have just built a $43 computer.

Somewhere in this whole process, Katie came up with the name “Red Shirt Linux.” And while it’s mostly going to be SuSE, Conectiva, and Mandrake, the name fits.

Preliminary tryouts look promising: All the hardware works, I was able to see old data on the hard disk before I repartitioned it, I could boot tomsrtbt off of a floppy and mess around under that. I tried Conectiva first, and it failed, but I think I’ve got a bad install CD. (The UML system I built from it has network problems, and the copy I installed on Ghostwheel is what trashed my partition table, so it doesn’t surprise me that it had problems here.) I’ll run a thorough memory test overnight just to make sure, but it looks like I’ve got a PC I can mess around with without risking any data!