BlackBerry Forums Support Community - View Single Post

vcoleman · 01-25-2006, 12:09 PM

Hi, I'm trying to put together a solid testing strategy for a BB application. It does not do any networking, except for invoking the browser to help pages and such. Right now the testing plan includes subscribing to a BES-HOST, and getting 5-6 different models with different Carriers. I've tried testing the APP with all the simulators but real BBs exhibit different behavior, such as timing for race-conditions, etc. Before embarking on a spending spree I'm wondering how other developers QA a BB app. Is a BES totally a waste (hosted option is pretty cheap) or does it present circumstances that a JDE with simulators can never replicate. What about different OS versions? Does QA involve compiling for different versions of the OS or just compiling once and running on several OS versions. What is customary in QA? Perhaps, it is common practice to test on just a couple of devices and cross your fingers... Do you have some general testing strategies that have worked?

Thanks,

Victor.

01-25-2006, 12:09 PM	#1
vcoleman New Member Join Date: May 2005 Model: 7290 Posts: 11	Solid Testing Strategy Please Login to Remove! Hi, I'm trying to put together a solid testing strategy for a BB application. It does not do any networking, except for invoking the browser to help pages and such. Right now the testing plan includes subscribing to a BES-HOST, and getting 5-6 different models with different Carriers. I've tried testing the APP with all the simulators but real BBs exhibit different behavior, such as timing for race-conditions, etc. Before embarking on a spending spree I'm wondering how other developers QA a BB app. Is a BES totally a waste (hosted option is pretty cheap) or does it present circumstances that a JDE with simulators can never replicate. What about different OS versions? Does QA involve compiling for different versions of the OS or just compiling once and running on several OS versions. What is customary in QA? Perhaps, it is common practice to test on just a couple of devices and cross your fingers... Do you have some general testing strategies that have worked? Thanks, Victor.
Offline