For my last blog post for this class, I found an article online that
talks about the practice of shifting-left in software quality assurance.
This approach more or less emphasizes the importance of introducing quality
assurance to earlier phases in the development process. Testing from the
initial phase of development is supposed to prevent the amount of defects
and issues from piling up at the end of development. Having testing done
throughout the development phases can also lessen the workload for the
quality assurance team.
According to the article, the cost of testing and post-production vastly
outweighs the cost of development and planning. It posits that testing
earlier and more frequently catches bugs earlier on, and reduces the overall
cost of development. This goes very hand in hand with the agile software
development methodology we learned about last semester. The world of
software development has become much more fast paced, and the current
landscape pushes for finished products with minimal defects at
launch.
I have seen online the mentality that a product that ships with any
problems is often ostracized. Consumers want minimal issues and problems
when interacting with any kind of software, and that goes doubly for large
companies. Having software testers involved since the start of development
would allow teams a more seamless development experience.
One model for development that the article proposes has each stage of
development separated by a quality check gate, in which test cases are
implemented. When all defects are found and fixed, only then can the
development team move on to the next stage of the process. I think this a
very good system that could fit well within the agile sprint methodology.
Leave time at the end of the sprint, but before the sprint retrospective,
for the quality assurance team to check the code, then at the retrospective
they can sign off on the state of the project. If there are any bugs that
could not be fixed within this sprint, the testers can assign it as an issue
for the next one.
During the Development Capstone project, this could be used to manage the
teams next semester. Have team members focus on quality assurance near the
end of the sprint, and then collect their feedback at the sprint
retrospective. Another way would be to have one or two teams act as quality
assurance throughout the whole semester. Either way it could save a bit of
headaches for everyone.
From the blog CS@Worcester Alejandro Professional Blog by amontesdeoca and used with permission of the author. All other rights reserved by the author.