Codermetrics

Name

Errors

Synopsis

Purpose

Measure the magnitude of production issues found related to areas that a coder is responsible for.

Formula

Errors = Sum (Severity for each Product Issues × Population Affected)

Example

Coder A has the following production issues found in the coder’s area of responsibility during two development iterations:

Iteration 1: Issue with Severity 2 affects 100% of users

Iteration 1: Issue with Severity 2 affects 50% of users

Iteration 2: Issue with Severity 3 affects 100% of users

Iteration 2: Issue with Severity 4 affects 100% of users

Iteration 2: Issue with Severity 3 affects 33% of users

You can calculate the Errors per iteration in the following way:

Errors Iteration 1 = (2 × 1.0) + (2 × 0.5) = 3

Errors Iteration 2 = (3 × 1.0) + (4 × 1.0) + (3 × 0.33) = 8

Then you can determine the Total and Average Errors:

Total Errors = (3 + 8) = 11

Average Errors = (3 + 8) / 2 = 5.5

Notes

Turnovers and Errors are the two basic metrics for coder mistakes. Errors are weighted by the severity of each issue and the percent of the customer population affected. The more severe the issue, and the more customers affected, the larger the assigned Error value.

You might wonder whether it is fair to weight Errors by the percent of the user population affected. Is a severe bug that affects all customers a bigger Error than a severe bug that only affects a few customers? In both cases, the coder made a mistake. And maybe the size of population affected is only greater based on the area of code where ...