Provided are responses to Philip's and Takuya's responses.
> is one hour enough preparation time? How do we determine that one hour is enough or not? We might correct questionnaire depending upon the amount of the review materials (i.e. number of class files) to adjust the enough-feeling one hour review though...
You would use Statistical Process Control.
> you might decide to not do the review meeting if there are no critical or major
> issues uncovered during preparation. You might decide to delay the group meeting if
> preparation time was not sufficient.
This may or may not be good because as you see, almost reviewers use the default severity (i.e. in our case, unset or normal) so that we can not see there exist really critical or major issues. Even though we can force them to chose one of them, it would be hard to determine that review should be held or not because the magnitude of severity that each reviewer thought might be different from that of the other reviewers think in the team phase.
You're right that we don't really use the severity. But, I think that is a problem that we have to fix either with Jupiter or in our process. My view is that severity is very important and it isn't something that we want to ignore.
> > - It would also be interesting to test the age old belief that reviews decrease the
> > number of defects. We can do this easily if we can associate a defect to a particular
> > class and checking if that class has been reviewed.
>
> I don't think this is easy at all to do. There are all sorts of conflicting independent
> variables, including the complexity of the code and the skill of the author. You'll need
> a very large sample size to factor this stuff out.
I agree with Philip. As well as the complexity and the skill, the deduction of defects might be caused by another factors such as test cases and so forth.
I don't understand both of your arguments. If we conduct a review and resolve the issues then aren't we saying (in the perfect world) that no other defects should occur?
thanks, aaron
