First, hello to Dan Frost. C.Douce wrote: > Whilst I do appreciate the simplicity (and elegance) of the > 'print "I'm here"' approach, when it comes to developing > embedded systems nothing can beat a monitor program, a > serial cable and the ability to plant break points in your > code when you begin to step into trouble.
I cannot foresee a time in which everyone will agree about debugging. While the embedded systems argument has been made many times, I think there still will be a segment that resist it. I am reminded of Linus Torvalds' long insistence of no official kernel debug interface (http://www.uwsg.iu.edu/hypermail/linux/kernel/0009.0/1148.html) Some interesting conjectures there about the impact of tools on how people think. Torvald's comments also remind me of a previous PPIG discuss thread which also explored the question of whether deeper understanding is hindered by the availability of a shallow-but-immediate feedback loop (http://www.mail-archive.com/discuss@ppig.org/msg00208.html).
On a more personal note I have trouble separating debugging from other programming activities, and thus reserve first dibs on withdrawing from debugging debates by claiming them to be battles between straw men. ;)
Regarding the earlier question of "standard diagnosis procedures", there may indeed be some fairly stable and general sets of patterns/strategies/plans/schemas/frames that debuggers try. For example, IIRC Weisers's slicing work arose from observations of (human) debuggers commonly performing backwards data flow tracing from the point at which actual program behaviour diverges from expected.
That said, I am still having a difficult time relating medical diagnosis procedures with programming simply because of the variety in programs. Perhaps experts on debugging distributed reactive systems are specialized in ways similar to dermatologists? Still, I know that studies have been done on medical problem analysis and I am intrigued as to whether anyone has specific ideas as to how they may be related.
Andrew ---------------------------------------------------------------------- PPIG Discuss List (discuss@ppig.org) Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss Announce admin: http://limitlessmail.net/mailman/listinfo/announce PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/
