December 05, 2017

Start with a plan
Software of any nontrivial size is complex, and fully testing every possible combination of inputs and validating the expected outputs is a luxury that few can afford. With a test plan, risk areas can be identified early on, and testing strategies can be identified to mitigate those risks. For example, you may decide that for unit testing, where an individual software function call is the unit under test, it is not necessary to test every value of an integer parameter. Instead you create a test strategy that says tests will be added for the maximum, mid and minimum values. In this way we cut the number of tests from 4,294,967,296 down to just 3. You can easily see that testing every possible combination of values for every parameter for every... Continue reading »

Posted by Jeffrey Fortin in Software Quality

December 01, 2017

Q4 is in full swing! With it comes its share of craziness. That is because some clients may need to proceed with last minute purchases while the yearly budget is in effect. For others, it is planning for next year's budget. Add team expansions, vacations (Diwali, Thanksgiving, Christmas...) and the development of cross-departmental links between the VectorCAST people and other parts of Vector Informatik in the mix, and it makes for a busy season. Clients need to be met, miles will be accumulated, crazy schedules will be observed. This is the time to show determination and stamina to end things on a good note!

Except that life happens. And that sometimes makes things, well, more sporty than usual.

Posted by Steve Barriault in Journal Entry

November 02, 2017

Code Correctness

Start with the requirements

Posted by Jeffrey Fortin in

September 22, 2017

You ran all of your required tests, and they all passed. But do you know how much of the code was even tested? If the only metric being evaluated is that the tests passed, then it’s possible to come away with the feeling that the product is in good shape and ready to release. But test results alone are often misleading.

Adding additional quality metrics such as test coverage, code complexity, total number of statements, etc. help by providing a more complete quality assessment. These metrics provide real information about your code and your tests.

Posted by Jeffrey Fortin in Embedded Testing

September 22, 2017

Being part of Vector Informatik means that Asia becomes an even more important part of our strategy in the world. Already, my territory has grown tremendously in the last few years, a reflection of the greater weight of the economic tigers that populate the area. It is certainly an exciting place to be these days, sometimes not unlike the Far West of yore - a tad wild, certainly ever-changing, but certainly fascinating and oh so much fun!

But then, getting more clients means expansion for my team, the International Manage Force (IMF). I already talked about how to recruit prospective team members in a... Continue reading »

Posted by Steve Barriault in Journal Entry

August 23, 2017

Regression Text

Every time we modify code we run the risk of introducing new bugs. When programmers make changes to software to fix errors that have been identified, it can cause functions that were previously working properly to experience new problems.

As a result, every change made during software development necessitates retesting in order to verify that the previous quality baseline is still intact. Regression testing, also known as verification testing, is the process of analyzing software after a change has been made to verify that the quality has not regressed. Regression testing is performed after a software update but also when the software is ported to a new runtime environment, a new compiler is used or when the software is migrated to a new hardware platform. 

Posted by Jeffrey Fortin in Software Quality