I come with a request of advices, hints, path to follow, etc.
The point is the following. My goal is to develop a QA/testing area into a software company whose product is a web application.
As usual there are a many functional testers that ensure the application has the right quality for being released to production.
My goal is to apply an strategy for helping the company to grow at the same time the number of testers do not do so exponentially. I mean if the functionality increases it is natural to increase the number of testers. Mainly I want to restrict the number of manual testers because I find it not as productive as it should be. You know one of the main human characteristic is that we commit mistakes.
So, the straightforward answer is to implement automation, I am already aware of some/many tools for doing so and we are already doing so. However, I wish to go even deeper. On the one hand side usually these tools are oriented to the final product but I wish to have a broader approach focused on the quality of each single artifact produced, in other words if each single building block has a high quality the probability of having a high quality on the system built by combining them is going to increase. So my goal is to apply QA to as many artifacts the different areas of the company produces, let say, technical design, implementation/code, may be architecture artifacts, etc. And of course to the whole product.
I have some ideas for some artifacts but no idea for some others. Any comment or criticism is going to be welcomed.
My plan is the following
Final product: default approach: manual plus automatic testing
Code: Unit testing combined with static analysis
Technical design: Formal specification languages such as TLA or event-B
Architecture artifats: No idea, my knowledge is quite short
Besides the skills needed for carrying it out, my question are the following:
- Do you think it is doable or realistic?
- Do you think this approach can help to improve the quality of the product?
- Is it worth?
- Tools to consider...
Thanks a lot in advance.
Aucun commentaire:
Enregistrer un commentaire