Stewart Stremler wrote: >.. >> As a programmer, all I can say is that if I had to _pay_ for my >> mistakes, then I wouldn't be without unit tests. > > You're penalizing the wrong level with that approach, if that's the > only change. You have to penalize management who cranks up the > schedule, and ladles "process" across everyone like it's a magic gravy, > and demands that quality improve as well.
No arguments from me about distribution and types of blame. And pardon me for trying too hard to inject some (probably irrelevant) imagery. But the truth is, I do have to pay (figuratively, if not in later repair effort) for my mistakes. One point of realism -- unit tests certainly can't protect against things you don't think to test (definition of bug?). But OTOH each new bug discovery can/should produce another test case. Regards, ..jim (got more OTOH's than any Hindu god) -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
