In Weinberg’s Why Software Gets In Trouble, he discusses the difference between failures and faults. Although an obviously old concept, I’m finding it to be a novel approach to thinking about these issues (maybe novel to just me).
What is a system, systems thinking, science, philosophy, theory? How are they related? How can they be used?
Is software testing just checking if software conforms to the requirements? Sometimes… but usually the answer is no.
Although the Octopus Deployment Framework is an awesome tool, it has a couple of flaws that frustrate me. Here are a few of the issues that I think would greatly improve the framework.
Octopus allows you to define both intra- and inter-server dependencies. This is one of the coolest things about Octopus. However, defining dependencies is not without problems.
The Octopus Deployment Framework is an internal Microsoft framework used for deployment on some Microsoft teams, but why is it useful?
Expected values are valuable in statistics and in software testing. However, they generally mean different things. How do they differ, and how can one be applied to the other?
The Monty Hall problem can be applied to software testing! The problem crystallizes a key component of software testing… asking the right questions.
Binomial coefficients can be used in software testing! They are already used in mathematics, statistics and combinatorics. Let’s see how they can be applied while testing software.
Assisted GPS is used by most mobile phone devices. It can help improve GPS in poor signal conditions. But what is it? How does it work?