Thread by @noahsussman

#testing concepts thread.

What is deduction?

What is induction?

What are the differences between deductive and inductive investigation of anomalies?

What is an anomaly?

How do anomalous events differ from “normal,” non-anomalous events?

How does #testing help?

If you can’t prove with confidence that testing will help, should you be performing testing at all in that case?

🦉

What is the cost of #testing?

“Root Cause” is debunked pseudoscience.

Avoid working for managers who think in terms of Root Cause.

Assigning blame for bugs in production has nothing to do with #testing.

Assigning blame at all, is a bad idea.

The following metrics have no intrinsic meaning and are only as useful as the context you provide for them:

• Passing/Failing Test Results
• Many/Few Lines Of Code
• Percentage Of Code Coverage
• High/Low Cyclomatic Complexity
• Passing/Failing Static Analysis
Results

Conquest’s Third Law:

The behavior of any bureaucratic organization can best be explained by assuming that it is controlled by a secret cabal of its enemies.

Murphy’s Law:

Whatever can go wrong will go wrong at the worst possible time.

Murphy’s Law “Classic:”

If there is a way to do something wrong then someone will do it.

Heinlein’s Razor:

Never assume malice where ignorance will suffice.

Corollary:

Sufficiently advance ignorance is indistinguishable from malice.

Cheops’ Law:

Nothing is ever delivered on time or under budget.

Corollary:

Unless you cook the books.

“The devil is in the details.”
— trad.

Hofstadter’s law:

It always takes longer than you expect, even when you take into account Hofstadter's Law.

There’s never time to do it right but there’s never time to do it over, either.

The following things are impossible:

• Complete Regression #Testing
• Bug-Free Software
• Fully Automated Testing

Corollary:

There will always be bugs in production no matter how much time and money you throw at the problem.

The quality of software product is not in how many bugs are found and fixed before release. It is in how the team responds to bugs found in production.

If a bug is found and fixed before any customer sees it then it wasn’t a bug it was an ad hoc code review the outcome of which was that a change was made 🤷🏻‍♀️

Bugs are emergent properties of software systems and thus cannot be prevented.

Change one and only one variable at a time.

For any software that is making money, no one person knows how it works.

Conway’s Law:

Software resembles the team that created it and vice versa.

Sometimes people are paid to behave in a way that you might consider negligent.

“Perverse Incentives” are rewards the team gives (intentionally or not) for actions that detract from software quality.

For instance a strong commitment to meeting deadlines creates a perverse incentive to minimize bugs that might interfere with the deadline.

Complete #testing is impossible therefore the more money you spend on QA, the lower the ROI.

.@allspaw convinced me of this so check out his Web Site!