Cutler is against simply committing the code and letting the testers figure out where the defects are; he says that you should do the best you possibly can before giving it to the testers.
Not surprisingly, that's exactly what Steve Maguire, who worked at Microsoft during the same period as Cutler, says in his book Writing Solid Code. In the late 80s and early 90s, a lot of programmers at Microsoft apparently didn't test their code much or at all, and both expected testers to identify their defects, and got mad at them when they did so, even though they were just shooting the messenger for reporting bugs that they caused.
However, I was under the impression that software development doesn't really have a dedicated "testing" division anymore. They just expect the programmers to write tests and find their bugs, and ideally, they'll be automated when possible.
I'm actually still in college, never having worked as a professional software developer, so I wouldn't know except secondhand.
1 comment
[ 4.3 ms ] story [ 17.0 ms ] threadNot surprisingly, that's exactly what Steve Maguire, who worked at Microsoft during the same period as Cutler, says in his book Writing Solid Code. In the late 80s and early 90s, a lot of programmers at Microsoft apparently didn't test their code much or at all, and both expected testers to identify their defects, and got mad at them when they did so, even though they were just shooting the messenger for reporting bugs that they caused.
However, I was under the impression that software development doesn't really have a dedicated "testing" division anymore. They just expect the programmers to write tests and find their bugs, and ideally, they'll be automated when possible.
I'm actually still in college, never having worked as a professional software developer, so I wouldn't know except secondhand.