Evaluation Of A Release Manager

1874 Words May 5th, 2015 8 Pages
Part A - Release Manager

In general we can model our system using the quality triangle shown in Figure 1.1 below. We can push the system in one or two dimensions by making concessions on the other dimensions. This suggests that to deliver on time with acceptable quality means we will have to sacrifice features and cost.

Figure 1.1
We can divert feature effort by delaying select features and moving development and testing efforts into other features to an extent. To do this we will list each major feature in each function area by its defect rate. We will then consider the importance of each feature to the customer. Any obviously less-desired, high-defect features should be delayed. We will have to consider the impact of any higher desired, high-defect features.

Effort dedicated to those features can be redirected to key features in the interim. In order to mitigate Brooke’s law (the idea that adding more people to a late project will delay, rather than expedite shipping), any diverted effort should be within the same functional group. If diverted testers and engineers have common domain knowledge with other features, we will be able to reduce the impact of task switching. After shipping, we can redirect the now-free developers and testers to the delayed features.

We can also strategically work with customers to help find defects and increase customer satisfaction. Our customer program group should increase beta enrollment in all industry sectors. We should identify key…
Open Document