Like everyone else with an Intel graphics chipset and some X bug at DebConf, I was happy to have Keith there to debug it. But I'd go one step futher, and recommend anyone who doesn't have an Intel graphics chipset and some X bug, to go find one of each (and Keith). It's worth the price of admission just to experience his debugging process.

If your experience is anything like mine, you'll have a lot of fun. You'll learn about some graphics-related thing, such as PCI configuration space and the LCD brightness register. And you might learn some new things about debugging. How twenty years of experience do come in handy at diagnosing what kind of crash happened, at a glance. That it's possible to still have fun working on something after those twenty years of experience. How to keep the bug submitter involved by letting them follow your thoughts as the problem is debugged, and letting them handle some of the skut-work. How to teach them what they need to do to handle the skut-work, without talking down, or wasting any time. How important it can be to have things like good hardware documentation. And probably lots more that I missed.