As Mor mentioned here a couple of weeks ago we recently conducted a "Blitz" against bugs.
Once in a while we try and allocate some time in order to solve all those neglected bugs that always get second priority. In addition, we also try to fix the bugs that were found in the application after a feature was launched. As I've mentioned before, the dynamic and intensive development environment does not always allow us to conduct a completely thorough QA prior to every upgrade.
In the past, we were able to allow ourselves to designate R&D time and effort for every Blitz; everyone stopped developing new features and concentrated solely on fixing bugs. This was not the case this time - besides working on the Blitz, there were 4 other projects in process for R&D to handle.
Another thing that was different in this Blitz was the fact that we now have 2 separate environments - one for Beta and one for Production. Not all bugs were supposed to be fixed in both environments, and it was a challenging task to make sure that each bug was not only fixed, but also fixed in the right place.
As soon as the developers started fixing bugs we began our race against the clock. We had to verify all the fixed bugs ASAP and make sure to reopen the problematic ones in time for the developers to re-fix them.
Because so many bugs were handled and therefore so many parts of the application were touched - a lot of time was invested in Sanity testing. The purpose of the Blitz is to stabilize the application, so long after the developers moved on to work on new projects, we found ourselves conducting more and more regression and sanity tests in order to make sure the fixes did not introduce new problems to the system.
Eventually, we made sure to fix all of the major bugs, and most importantly, those bugs that affect users the most. As a result, I can positively say that our application is even more stable and user friendly today.
Comments