Information Technology Reference
In-Depth Information
outside of management. Sommerville cites the ACM/IEEE-CS joint task force in
making it clear that the competence and personal character of colleagues should be
respected (Sommerville
2004
). To do so, it is important to leave out personal
information of the developers when communicating information to people that is
not their management.
Beside not revealing personal performance information publically there are a
few other etiquette matters that should be considered. First, metrics should be
considered with sensitivity. Even if personal information is not included some
developers may find that they feel targeted and slighted by the presentation of the
results. The results of the tests and analysis should not become obsessed with a
particular area that seems to be lacking or one portion that seems to be flourishing.
In no circumstances should individuals or teams be either, praised or threatened.
This is not the point of such measurements. Instead use managerial channels to
asses a team or individual's performance (Pressman
2005
).
Always keep in mind that less than satisfactory measurements of performance
do not necessarily denote less the satisfactory effort or ability. A high number of
faults from a certain module during testing may not indicate that the implemen-
tation was poor, but maybe the design or theory was not well constructed. Anytime
an abnormally good or poor measurement is observed it is prudent to revisit the
data collection and analysis of said attribute or even the meta-metric itself.
12.1.4 Private Versus Public Metrics
Metric data will be collected and stored, but not all of this data is fit to be
distributed to the public at large. Some of the data collected will be of a nature
sensitive to individual teams or developers and should be distributed only to
management or quality control teams within the organization. This information
can be a strong driver when presented to the team or individual. It will allow them
to see the baseline created by their co-workers and where they stand in relation to
it. Setting goals and demonstrating progress will drive workers to increase their
quality and quantity of work (Pressman
2005
).
Some measurements will be worth promoting and publishing. Metrics that
indicate performance and reliability of the product seem the obvious choice when
such measures are compared to the competition. This will often ''prove'' some
kind of advantage. The measures using collective and assimilated information
should be used with caution as they can be misleading and easily misused.
12.1.5 Baseline Measurements
A baseline measurement is one that can be compiled from past projects. Baseline
measurements provide a sense of where you are in relation to other projects. Using
