Plenty of things went wrong in lab class today, most of it probably my fault. You see, I have 29 students in Operating Systems, and last week, I introduced them to the Linux kernel. If you've ever had to configure and build a kernel, you can probably tell what kind of mess I was headed in: hundreds of complicated options, long compilation times, and the high probability that your operating system won't boot.
As we say: nosebleed time.
In hindsight, things could have gone better if I had written the lab manual more carefully and tested each and every step. But I felt like going cowboy with this one.
On the other hand, I'm of the firm opinion that you learn more when things go wrong than when they go right. Next time, you know what NOT to do, and that's important, too. No better place to go wrong than in the lab, where nothing will explode, no accounts receivables systems will go down, and no one will get seriously injured. Or so we hope.
I think in the long run students will respect a teacher more if they can enter a less-than-ideal situation, admit the boo-boos, and figure a way out of that logically. That's what I'm hoping, anyhow.
And yes, I will fix up that danged kernel configuration lab manual.