My experience with the Agile process has been geared to
manual and automated testing. The
software developers spend 3 sprint periods developing new features and fixing
defects. At the end of the 3 sprints the
developers deliver a new product with an outline of new features and fixed
defects. This gives the testing team 2
sprints to test the software and report any new defects or report if existing
defects still exist.
For this process to work correctly the developers must be
able to deliver working software by the end of the period and when the testers receive
the software they must have all resources needed in a properly operational
state. In my experience it has been very
difficult for developers to release working software on time which limits the
time available for testers.
Once the testers receive the software there are sometimes a
shortage of resource to do the proper testing.
Many times this is because the testing resources are expensive. There
often are technical or mechanical issues that make some testing resources
unavailable.
After the testing has been done there are new defects that
must be planned into the next few sprints.
When the software has many new defects this limits the amount of new
features that the developers can implement within the next few sprints. With an unknown amount of defects after each
testing sprint it makes it hard to plan new features.
In summary I believe that to create an environment that is
conducive to proper testing the organization must have very good procedures in
place to ensure that deadlines are met and resources are available when needed.
Again, this is another example of how it takes an organization some trial and
error to properly use the Agile development process.
Leslie P ---Extra Blog Post