Friday, May 20, 2005

When is a project ready for release?

http://www.testingeducation.org/a/TheOngoingRevolution.pdf

A project is complete enough to release when it provides enough of the features, delivers enough of the benefits (the features have to work well enough together for the user to actually succeed in using the product to get real work done), is documented well enough for the user, validated well enough for regulators or other stakeholders (e.g. litigators of the future) who have a legitimate interest in the validation, has been sufficiently instrumented, documented, and troubleshot to be ready for field or phone support, is sufficiently ready for maintenance, localization or porting to the next environment (readiness might include having maintainability features in the code as well as effective version control and other maintainability-enhancing development processes in place), is acceptable to the key stakeholders, and has few enough bugs. This list is not exhaustive, but it illustrates the multidimensionality of the release decision. Many companies appraise status and make release decisions in the context of project team meetings, with representatives of all of the different workgroups involved in the project. They wouldn't need these team meetings if the status and release information were one-dimensional (bug counts). We describe these dimensions in the language of "good
enough" because projects differ in their fluidity. One organization might insist on coding everything agreed to in a requirements specification but do little or nothing to enable later modification. Another might emphasize high reliability and be willing to release a product with fewer than the desired number of features so long as
the ones that are included all work well. Even if we restrict our focus to bugs, the critical question is not how many bugs are in the product, nor how many bugs can be found in testing but is instead how reliable the product will be in the field (Musa, 1999), for example how many bugs will be encountered in the field, how often, by how many people, and how costly they will be. (Kaner & Bond, 2004)
Post a Comment