News for hackers. Stuff that matters.

A comment on I ruin developers’ lives with my code reviews and I'm sorry. Parent comment

Re:I ruin developers’ lives with my code reviews and I'm sorry by btilly on Monday February 18, 2019 @03:04PM (#19193884)
The issue is that code reviews are hard on people and good on code bases.

Dollar for dollar, code reviews are more effective at finding and fixing bugs than any other activity we can do, including QA tests. However code reviews are very hard on people, and can easily create conflict.

Which one matters more, and how careful people are, varies widely by organization.

[ Parent | on HN ]
Re:I ruin developers’ lives with my code reviews and I'm sorry by jacques_chester on Monday February 18, 2019 @05:25PM (#19195054)
My recollection is that the studies showing that code reviews have high defect yield are for extremely strict methods. Everyone in a room, prescribed roles for each person, line by line, checklists, forms, the whole nine yards.

Comments on pull requests are not on the same level and those findings shouldn't be hastily generalised.

[ Parent | on HN ]
Re:I ruin developers’ lives with my code reviews and I'm sorry by btilly on Monday February 18, 2019 @05:49PM (#19195202)
My impression is that comments on pull requests do not catch as many defects as formal review, but catch even more per dollar spent.
[ Parent | on HN ]