Home > Articles > Software Development & Management

  • Print
  • + Share This
This chapter is from the book

Guiding Principles for Reviews

All peer reviews should follow several basic principles to make them powerful contributors to product quality and team culture (Wiegers 1996a). First, check your egos at the door. There are two aspects to being egoless. As an author, keep an open mind and be receptive to suggestions for improvements. Avoid the temptation to argue every point raised, defend your decisions to the death, or explain away errors. As a reviewer, remember that you're not trying to show how much smarter you are than the author.

Another useful guideline is to keep the review team small, generally between three and seven participants. Larger teams make the review more expensive without adding proportionate value. They slow the rate at which the group can cover material in a review meeting. Large groups are prone to distracting side conversations and can easily go off on time-wasting tangents. Chapter 12 suggests what to do if a lot of people wish to participate in a review.

The prime objective of a peer review is defect detection, so strive to find problems during reviews, but don't try to solve them. Technical people like to tackle challenging problems; that's why we're in the software business. However, the author should fix the identified problems after the review meeting, working with selected reviewers on specific issues to reap the benefits of another's experience. Formal reviews, such as inspections, include a moderator or review leader role. As described in Chapter 5, the inspection moderator is responsible for keeping the meeting focused on finding defects and for limiting problem-solving discussions to just a minute or two. If you don't use a moderator, the participants will have to monitor themselves so the meeting doesn't derail into an extended brainstorming session on the first bug found.

Another guiding principle is to limit review meetings to about two hours. My friend Matt once pointed out to me that "the mind cannot absorb what the body cannot endure." When you are distracted by physical discomfort or exhaustion, you're no longer an effective reviewer. Take a short break halfway through a long review meeting, and come back tomorrow to finish if you didn't properly cover all the material in the first meeting. Scope the work into logical chunks that the team can examine in one to two hours, based on your organization's historical rates for reviewing or inspecting different types of work products (see Chapter 5).

To use the team's time as efficiently as possible, require advance preparation for formal reviews. During informal review meetings, participants come in cold and listen to the author describe his work. Serious defect-hunting efforts, such as inspection, demand that the meeting participants have already examined the product on their own to understand it and find issues to raise during the meeting. The participants need to receive the review materials several days prior to the meeting to give them time to prepare for the inspection meeting.

Being sensitive to the human and cultural issues of peer reviews and following these basic guidelines will maximize the contribution reviews make to your development and maintenance projects.

  • + Share This
  • 🔖 Save To Your Account