Welcome to the 82nd issue!

A few months ago, I mentioned the article about how Facebook is finding and fixing regressions faster stating that the later the bug is found, the costlier it is to fix.

If Google, Facebook and some research papers say so, why wouldn't it be?

Well, that's where it gets funny.

As it turns out, the most-quoted research about the cost of a software bug doesn't even exist! And there's more. Hillel Wayne explains that many others are full of bad assumptions or errors, too.

So, how is it? Are the late-spotted bugs costlier to fix?

It depends.

We've learned it's hard to prove or deny scientifically. But I still think testers should be involved in the early phases of the software development life cycle.

Spotting an invalid requirement when it's still on paper is — and always will be — cheaper to fix (in terms of both time and cost) compared to finding out about it when it's already delivered.

This is common sense. We don't need research papers for that.

Happy testing!

Dawid Dylowicz

You might have heard that detailed test cases are a waste of time. Why? Brijesh Deb provides ten solid reasons for that.

And here's another one from Maaret Pyhäjärvi who shows The Most Overused Test Example — Login.

Many teams claim they do agile testing but not many do it right. Blake Norrish put together a really good list of common anti-patterns that occur fairly often (unfortunately).

I love the way Maaret Pyhäjärvi explained the relationship with testing varying on the point of view. So simple and so true.

Additionally, James Thomas gives a great example of how to ask the right questions in Mass Testing.

How important is communication for software testers? Very. Sérgio Martins gives a few valuable pieces of advice that can help you win in various situations.

Speaking of which, Edib Imamović explains how to Report a bug like you mean it! in a cool way.

This is an amazing, thought-provoking article from Daniel Terhorst-North — the inventor of BDD with 30 years of experience in software development — discussing what testing is, what it isn't, and what practices such as TDD, BDD and ATDD have to do with testing.

Two weeks ago, I linked to the State of Testing in DevOps report. This inspired Beth Marshall to analyse and share what she learned from it. Thanks!

Similarly, Bridget Hughes shared some insights about The Changing Role of Software Testing based on the same report.

Mark Walsh correctly points out that many engineers still don't write tests for production code. And they find plenty of excuses for that, all of which Mark debunks here.

Sounds like yes is the obvious answer? Well, Alan Richardson might surprise you in this case. And I agree with him entirely.

This is a very detailed guide to REST API testing by Abhishek Chauhan, including plenty of examples.

And for an equally comprehensive read, have a look at Handling REST APIs by Pricilla Bilavendran.

Interested in testing serverless apps? Patrick Döring shares code examples showing how to test them using two open-source tools: 7Facette and Playwright.

And for more about serverless testing, check out Andy Blackledge's latest article on Step Function integration testing with CDK.

A good test doesn't only check for the regression bugs but also acts as a form of feature documentation. This is why it's important to do the naming well, as Dawid Sibiński explains.

If you're just starting with Cypress (or a similar web test automation tool), here are the key things to look out for. By Ashutosh Bhardwaj.

Moreover, Gleb Bahmutov shows a few examples of how to Avoid Cypress Pyramid of Doom.

Danny Shain did a big favour to all Appium fans here by putting together this list of articles and guides on writing effective mobile tests.

And if you want to learn about other mobile testing frameworks, Archana Jayappa and Lalitanand Dandge listed the pros and cons of the five most popular ones.

Kislay Verma shared a nice summary of each chapter from the popular DevOps book: Accelerate: Building and Scaling High-Performing Technology Organizations.

Here's a nice overview by Oleksandr Romanov of the Designing Data-Intensive Applications book about data storage, delivery and distributed systems.

It's not all about automation, as Pramod Dutta explains in this short video, sharing a bunch of valuable tips on what manual testers can do to stay relevant in their field.

Referring to the theme of today's issue, here's a 2-minute video presenting Gil Zilberfeld's opinion on the cost of bugs. Good timing, Gil!

Some bugs are features... 😂

Thanks for reading!

If you like this newsletter and it helps you become a better tester, you can say thanks and buy me a coffee.

Dawid Dylowicz