I have this mish-mash of Python skits playing in my head:
What is your problem?
Well, rebooting to apply updates and patches to your OS is a pain, and in some cases it’s actually really bad.
What is your aim?
To do something about the problem! I guess dynamic update has a good kind of ring to it.
…and what is its scope?
I don’t have time to rewrite the OS, or write my own OS, and other people have done lots of stuff in user-level updates already, so I suppose that limits the scope.
What are your hypotheses?
What is your research method?
To hack something up, perform some experiments, and at the end of it all try to write a coherent thesis about it.
What are your results?
I don’t know, I have some numbers, does that count?
What do you conclude?
I conclude that it’s possible for my toy research system, no, wait, I conclude that it’s possible for all sorts of systems if you believe my very convincing arguments at generalisation, which are… um… Aaaargh!
(falls into the chasm)